|Publication number||US8183593 B2|
|Application number||US 12/580,658|
|Publication date||May 22, 2012|
|Filing date||Oct 16, 2009|
|Priority date||Oct 16, 2009|
|Also published as||US8482072, US20110089540, US20120229941|
|Publication number||12580658, 580658, US 8183593 B2, US 8183593B2, US-B2-8183593, US8183593 B2, US8183593B2|
|Inventors||Robert J. Drost, Robert D. Hopkins, Alex Chow|
|Original Assignee||Oracle America, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (13), Non-Patent Citations (3), Referenced by (2), Classifications (25), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present disclosure relates to a semiconductor die that includes an integrated electro-static discharge (ESD) device to protect the semiconductor die from transient ESD signals.
2. Related Art
ESD events commonly arise during the manufacture, assembly or use of systems containing integrated circuits. While the amount of energy dissipated during an ESD event may be comparatively small on a human scale (akin to the shock you receive when touching a metal object after walking across a thick carpet), from the perspective of the small, sensitive electronics on an integrated circuit, the destructive power can be comparable to a lightning strike, and can cause significant damage to integrated circuits.
A variety of existing ESD-protection circuits and techniques are used to protect integrated circuits from ESD events. For example, one existing ESD-protection circuit includes reverse-biased metal-oxide-semiconductor (MOS) or PN diodes that turn on during an ESD event. However, when sized to safely absorb an ESD event, this ESD-protection circuit can significantly increase the parasitic capacitance of the input/output (I/O) pads in an integrated circuit. This increase in parasitic capacitance is problematic because it can: reduce edge rates by introducing inter-symbol-interference (ISI); cause reflections; and cause near-end or far-end crosstalk. These problems can significantly reduce the maximum data rate that can be communicated through I/O channels. Furthermore, parasitic capacitance consumes power as signals charge and discharge the additional load.
Additionally, the parasitic capacitance introduced by existing ESD-protection circuits (such as MOS transistors, PN junctions, or N-well resistors) is typically a nonlinear function of the transient ESD signal amplitude. This nonlinearity further deteriorates the I/O pad's termination properties for high-speed signaling, and introduces distortion in analog-to-digital (A2D) converters that limits their conversion accuracy.
Hence, what is needed is an ESD-protection circuit without the problems described above.
One embodiment of the present disclosure provides a semiconductor die that includes: a substrate, a first layer deposited on the substrate, and a second layer deposited on the first layer. The first layer includes an ESD device with at least a metal component coupled to an I/O pad, and coupled to ground via a signal line. Moreover, adjacent edges of the metal component and the I/O pad are separated by a spacing that defines an ESD gap. When a field-emission or ionization current flows across the ESD gap, the metal component provides a discharge path to ground for transient ESD signals. Furthermore, the second layer at least partially encloses the ESD gap between the metal component and the I/O pad, thereby providing a gas in the ESD gap.
In some embodiments, a surface of the second layer and a surface of the substrate define a cavity that fully encloses the ESD gap. Alternatively, the second layer may include an opening, defined in part by a surface of the second layer, which exposes the ESD gap to the gas. Furthermore, the gas in the cavity may be air or other than air.
Note that at least one of the adjacent edges may have an arrow shape in a plane of the first layer. In some embodiments, the adjacent edges both have an arrow shape in a plane of the first layer, and the spacing that defines the ESD gap is between tips of the arrow-shaped adjacent edges.
In some embodiments, the semiconductor die includes a third layer between the substrate and the first layer which is underneath at least the ESD gap. For example, the third layer may include a dielectric. This dielectric and/or the spacing may be used to select a turn-on voltage of the ESD device during design.
Additionally, the semiconductor die may include multiple ESD devices in the first layer, where the metal component in a given one of the ESD devices is coupled to a corresponding I/O pad, and is coupled to ground via the signal line. For example, the ESD devices may be arranged in: a 1-dimensional array along the signal line in the first layer or a 2-dimensional array in the first layer. Furthermore, the turn-on voltage of the given ESD device may be in one of at least two ranges of voltages, which are separated by a threshold voltage. Note that a size of the given ESD device may correspond to the turn-on voltage.
In some embodiments, the semiconductor die includes an ESD diode, which is coupled in parallel with the ESD device. For example, a low-pass filter may couple the ESD device and the ESD diode. Note that the ESD device may have a lower capacitance than the ESD diode.
Another embodiment provides a chip package that includes the semiconductor die.
Another embodiment provides an electronic device that includes the chip package.
Note that like reference numerals refer to corresponding parts throughout the drawings. Moreover, multiple instances of the same part are designated by a common prefix separated from an instance number by a dash.
The following description is presented to enable any person skilled in the art to make and use the disclosure, and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
Embodiments of a semiconductor die, a chip package that includes the semiconductor die, and an electronic device that includes the chip package are described. This semiconductor die includes an ESD device with a metal component coupled to an I/O pad, and coupled to ground via a signal line. Moreover, adjacent edges of the metal component and the I/O pad are separated by a spacing that defines an ESD gap. When a field-emission or ionization current flows across the ESD gap, the metal component provides a discharge path to ground for transient ESD signals. Furthermore, the ESD gap is at least partially enclosed so that there is gas in the ESD gap.
Relative to existing ESD-protection circuits, the ESD device: absorbs or dissipates more energy, has less parasitic capacitance, and has a more linear response. Consequently, the ESD device provides improved protection from ESD events, and offers improved performance in high-speed circuits or in circuits with precision signals.
We now describe embodiments of a semiconductor die. An ESD event has two locations on a chip or semiconductor die: one into which an ESD current enters the semiconductor die, and one from which the current leaves the semiconductor die. For most semiconductor dies, the wires and circuits are covered with a protective insulating layer called the passivation. Typically, only the I/O pads of the semiconductor die are exposed to air by etching an opening in the passivation over the I/O pads. In order to protect the wires and circuits on the semiconductor die, ESD-protection circuits provide an alternative path for the ESD current between any pair of I/O pads.
One configuration for protecting a semiconductor die 110-1 is shown in
As described further below with reference to
Note that the metal path in ESD devices 118 may be used for another purpose as well (such as a power, ground or signal rail) because, during normal operation, the ESD gaps in all of ESD devices 118 are insulating. As shown in
We now describe a number of ways to construct the ESD gap in ESD devices 118.
In some embodiments, at least one of adjacent edges 212 may have an arrow shape, such as a sharp tip. This is shown in
In some embodiments, adjacent edges 212 both have an arrow shape in a plane of metal component 210-1, and spacing 214, which defines the ESD gap, is between tips of the arrow-shaped adjacent edges. This is shown
In some embodiments, an ESD device includes a number of paired sharp tips. This is shown in
While several of the preceding embodiments have included a lithographically defined sharp tip, in some embodiments one or more corners of rectangular I/O pad 116-1 and/or metal component 210-1 are used to provide one or more sharp tips. This is shown in
We now further describe the structure in embodiments of an ESD device.
Note that the metal in the ESD device on both metal component 210-1 and I/O-pad 116-1 sides of the ESD gap are exposed to a gas (such as air) so that the ESD gap will break down at the dielectric breakdown voltage of the gas. This is useful because the dielectric strength of solid insulators (such as silicon dioxide) is typically thousands of times higher per micron than a gas, and field emission is strongest in a gas. As a consequence, without the exposed ESD gap, the ESD device would not activate for many ESD events.
In some embodiments, a surface of layer 314 and a surface of substrate 310 define a cavity that fully encloses the ESD gap. This is shown in
In some semiconductor fabrication processes, having an open passivation cut across an edge of a top layer metal or having a large top-layer metal edge-to-edge spacing (for example, 5 μm) may not be allowed. This problem may be overcome by requesting exemptions for the passivation-cut restrictions and/or by requesting smaller edge-to-edge spacings. Alternatively, the so-called ‘fuse’ layout structure, which is typically available in many semiconductor processes, may be used. In a fuse layout structure, a lower layer of metal (typically one or two layers down from the top layer) is exposed by etching the passivation layers in addition to the oxide layers covering the fuse metal. This may be useful because the fuse layout rules typically allow for smaller gaps between metal wires than the top layer metal. A cross-sectional view of such a semiconductor die 340 is shown in
Note that an ESD event may have a wide range of possible voltages and energies. Clearly, protecting a semiconductor die from the highest voltage and energy is desirable, but this is usually traded off against the cost of that protection. The cost comes in the form of the area of the ESD devices, their structures and the associated parasitic capacitance (and, hence, the bandwidth reduction and power consumption). A number of standard ESD events are defined that provide useful protection goals without too much cost. Two example ESD events are the machine model (MM) and human body model (HBM). A MM ESD event is a 200 pF capacitor, charged to a few hundred volts to a few kilo-volts, which is instantaneously coupled to the I/O pads in series with a 750 nH inductor and a 1Ω resistor. The human body model is a 100 pF capacitor, charged to 1 to 10 kV, which is instantaneously coupled to the I/O pad in series with a 7.5 uH inductor and a 1.5 kΩ resistor. Because of the significantly higher resistance and inductance in the MM, ESD-protection circuits typically withstand about a ten-times higher ESD charging voltage in this ESD event.
To address the range of possible voltages and energies in ESD events, and given the layout constraints and the desired protection levels, a semiconductor die may include different ESD devices for some of the I/O pads. Furthermore, if a supply rail is close to a group of I/O pads, and a different supply rail is close to other I/O pads, the supply rails may have additional ESD device(s) between them that provide a transient discharge path for I/O pads that have their protection via different supply rails. Note that in these embodiments there may be more than two ESD devices in series in the discharge path between any pair of I/O pads. Consequently, the effective pad-to-pad voltage or turn-on voltage of the ESD gaps in the two or more ESD devices may be increased, which may reduce the ESD protection for these I/O pads. This may be mitigated by selecting a different turn-on voltage for these ESD devices during design.
A variety of techniques may be used to select the turn-on voltage of a given ESD device. For example, the turn-on voltage of a given ESD device may be selected or modified by including an optional layer 316 (such as a dielectric) in semiconductor dies 300 (
We now describe ESD-protection circuits that include one or more ESD devices.
An example of such an ESD-protection approach is shown in
The intent of the primary ESD structure is to reduce a sub-nanosecond multi-thousand-volt ESD event on I/O pad 116-1 to a few hundred volts at resistor 510. Moreover, the low-pass filter slows the edge of the ESD event to over a number of nanoseconds. Finally the intent of the secondary ESD structure is to prevent sensitive on-chip circuits (such as circuit 514) from experiencing more than a few volts. Note after passing through ESD device 118-1 and/or diode 512-1, the ESD current may be electrically coupled to ground via transistor 516 or capacitor 518. Furthermore, note that a given ESD device (such as ESD device 118-1) may have a lower capacitance than a given diode (such as diode 512-1).
Given this division of labor, the primary ESD structure dissipates most of the ESD event energy. Consequently, in existing ESD-protection circuits the primary ESD structure has the largest cost in terms of area and parasitic capacitance. By replacing the semiconductor devices (such as diodes) in the primary ESD structure with ESD devices 118, ESD-protection circuit 500 may offer improved protection with reduced parasitic capacitance.
Note that Paschen's curve shows that the gaseous-breakdown effect in air corresponds to a breakdown voltage of around 330 V plus 3 V/βm for the ESD gap. However, this curve is only valid for air gaps larger than 5 μm. The lesser known modified Paschen's curve indicates that, for air gaps smaller than 5 μm, the adjacent metal edges begin to conduct via the field-emission effect. This effect has a smoother turn-on transient than gaseous breakdown of air. For these small gaps, the field emission turns on and conducts significant current for a threshold voltage of approximately 75 V/μm for the ESD gap. Therefore, if the spacing of the ESD gap in an ESD device is kept to under a few microns, it can serve the function of the primary ESD structure and can reduce a multi-thousand-volt ESD event to a few hundred volts. Because it may be difficult to reduce the ESD event to a few volts, given typical minimum lithographic metal-to-metal spacing in high-layer metals, ESD-protection circuit 500 may include the low-pass filter and the secondary ESD structure for further reduction to a few volts. However, as noted previously, the turn-on voltage of a given ESD device may be selected during design. If the resulting turn-on voltage is small enough, the low-pass filter and/or the secondary ESD structure may be removed.
The semiconductor dies in the preceding embodiments may be included in corresponding chip packages. A given chip package may include: a ceramic housing, encapsulation, and connectors or pins for electrical coupling to other components. Furthermore, the resulting chips may include: an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a processor, memory, graphics, a switch, and/or an integrated circuit that includes an I/O circuit.
We now describe embodiments of an electronic device and a computer system.
Memory 724 in the device 700 may include volatile memory and/or non-volatile memory. More specifically, memory 724 may include: ROM, RAM, EPROM, EEPROM, flash, one or more smart cards, one or more magnetic disc storage devices, and/or one or more optical storage devices. Memory 724 may store an operating system 726 that includes procedures (or a set of instructions) for handling various basic system services for performing hardware-dependent tasks. Moreover, memory 724 may also store communications procedures (or a set of instructions) in a communication module 728. These communication procedures may be used for communicating with one or more computers, devices and/or servers, including computers, devices and/or servers that are remotely located with respect to the device 700.
Memory 724 may also include one or more program modules 730 (or a set of instructions). Note that one or more of program modules 730 may constitute a computer-program mechanism. Instructions in the various modules in the memory 724 may be implemented in: a high-level procedural language, an object-oriented programming language, and/or in an assembly or machine language. The programming language may be compiled or interpreted, i.e., configurable or configured, to be executed by the one or more processors (or processor cores) 710.
Computer system 700 may include, but is not limited to: a server, a laptop computer, a personal computer, a work station, a mainframe computer, a blade, an enterprise computer, a data center, a portable-computing device, a supercomputer, a network-attached-storage (NAS) system, a storage-area-network (SAN) system, and/or another electronic computing device. Note that computer system 700 may be at one location or may be distributed over multiple, geographically dispersed locations.
Semiconductor die 110-1 (
Note that some or all of the functionality of electronic device 600 (
While the preceding embodiments uses semiconductor dies 110 (such as silicon), in other embodiments a different material than a semiconductor may be used as the substrate material.
The foregoing descriptions of embodiments of the present disclosure have been presented for purposes of illustration and description only. They are not intended to be exhaustive or to limit the present disclosure to the forms disclosed. Accordingly, many modifications and variations will be apparent to practitioners skilled in the art. Additionally, the above disclosure is not intended to limit the present disclosure. The scope of the present disclosure is defined by the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5731223 *||Sep 24, 1996||Mar 24, 1998||Lsi Logic Corporation||Array of solder pads on an integrated circuit|
|US6117299 *||May 9, 1997||Sep 12, 2000||Mcnc||Methods of electroplating solder bumps of uniform height on integrated circuit substrates|
|US6610262 *||Mar 4, 2002||Aug 26, 2003||Taiwan Semiconductor Manufacturing Company||Depletion mode SCR for low capacitance ESD input protection|
|US6682993 *||May 31, 2002||Jan 27, 2004||Taiwan Semiconductor Manufacturing Company||Effective Vcc to Vss power ESD protection device|
|US7903379 *||Jul 31, 2007||Mar 8, 2011||Micron Technology, Inc.||Cascode I/O driver with improved ESD operation|
|US7973365 *||Jan 25, 2008||Jul 5, 2011||Infineon Technologies Ag||Integrated RF ESD protection for high frequency circuits|
|US8080442 *||Jun 21, 2008||Dec 20, 2011||Elm Technology Corporation||Vertical system integration|
|US20040089909 *||Nov 7, 2002||May 13, 2004||Taiwan Semiconductor Manufacturing Co., Ltd.||Integrated circuit having improved esd protection|
|US20050247979 *||Apr 27, 2005||Nov 10, 2005||Greg Fung||ESD protection structure with SiGe BJT devices|
|US20060065931 *||Sep 27, 2004||Mar 30, 2006||Taiwan Semiconductor Manufacturing Company, Ltd.||ESD protection for high voltage applications|
|US20090111235 *||Jan 5, 2009||Apr 30, 2009||Lawrence Clevenger||Semiconductor Integrated Circuit Devices Having High-Q Wafer Back-Side Capacitors|
|US20090284883 *||May 19, 2009||Nov 19, 2009||Samsung Electronics Co., Ltd||Electronic device having electrostatic discharge protection device and methods of fabricating the same|
|US20100187652 *||Jul 28, 2009||Jul 29, 2010||Yang Xiao Charles||Method and structures of monolithically integrated esd suppression device|
|1||Chun, Jung-Noon et al., "EDS Protection Circuits for Advanced CMOS Technologies", Stanford University, Jun. 2006.|
|2||Kelly, M.A. et al., "An Investigation of Human Body Electrostatic Discharge", ISTFA '93: The 19th International Symposium for Testing & Failure Analysis, Los Angeles , California, USA/Nov. 15-19, 1993, pp. 167-173.|
|3||Voldman, S. et al., "A Strategy for Characterization and Evaluation of ESD Robustness of CMOS Semiconductor Technologies", SEMATECH ESD Technology Working Group, EOS/ESD Symposium 99-212, IEEE May 2009, pp. 3A.4.1-3A.4.13.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8716843 *||Apr 25, 2012||May 6, 2014||Stmicroelectronics S.A.||Microelectronic chip, component containing such a chip and manufacturing method|
|US20120273952 *||Apr 25, 2012||Nov 1, 2012||Stmicroelectronics (Crolles 2) Sas||Microelectronic chip, component containing such a chip and manufacturing method|
|U.S. Classification||257/173, 257/355, 257/660, 257/174, 257/659, 257/E23.114, 257/E23.115|
|International Classification||H01L29/73, H01L23/552, H01L31/111, H01L29/72, H01L23/62, H01L29/74|
|Cooperative Classification||H01L2924/14, H01L2924/13091, H01L2224/02166, H01L2924/00014, H01L2924/0002, H01L2924/1305, H01L27/0288, H01L23/60, H01L24/05, H01L2224/05567|
|European Classification||H01L27/02B4F8, H01L23/60|
|Nov 10, 2009||AS||Assignment|
Owner name: SUN MICROSYSTEMS, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:DROST, ROBERT J.;HOPKINS, ROBERT D.;CHOW, ALEX;SIGNING DATES FROM 20090910 TO 20091009;REEL/FRAME:023497/0879
|Nov 4, 2015||FPAY||Fee payment|
Year of fee payment: 4
|Dec 16, 2015||AS||Assignment|
Owner name: ORACLE AMERICA, INC., CALIFORNIA
Free format text: MERGER AND CHANGE OF NAME;ASSIGNORS:ORACLE USA, INC.;SUN MICROSYSTEMS, INC.;ORACLE AMERICA, INC.;REEL/FRAME:037311/0171
Effective date: 20100212